Harvard Reports on Cancer Prevention
Volume I: Human Causes of Cancer
Cancer Causes & Control:
An International Journal of Studies of Cancer in Human Populations
Official Journal of the International Association of Cancer Registries
Volume 7 Supplement November 1996 ISSN 0957-5243

Occupation

Introduction
Reviews conducted since 1970 by the International Agency for Research on Cancer (IARC) indicate that most of the exposures that have been judged to cause cancer in humans have occurred in the workplace.¹ In the workplace, people are likely to be exposed to a broad spectrum of carcinogens at high concentrations; therefore, epidemiologists and physicians are more likely to identify these cancer-causing agents. In contrast, it has been estimated that less than five percent of cancer (and at most 15 percent) in men and one percent of cancer in women (and no more than five percent) can be attributed to occupational factors.² This apparent discrepancy relates to the quantity and quality of information available on human carcinogens as well as the fact that most of us have minimal exposure to potentially toxic chemicals in the workplace.

Clearly, tobacco smoke is the single factor known to have caused the highest proportion of cancer. Next, a wide variety of exposures to chemicals in the workplace have been identified as causing a variety of cancers. Tobacco and many workplace exposures are relatively strong causes of cancer. The rate of cancer among people exposed to these agents has been 10 to 100 times the rate in unexposed people. Other causes of cancer in humans, while possibly present among a larger proportion of the population, are likely to be weak and thus difficult to identify.

Evaluation of exposures
The designation of a substance as a human carcinogen is a matter of collective judgment. The reviews conducted by IARC are based on a comprehensive evaluation of the current scientific literature relating to chemical, physical, or biological exposures in humans and other animals. Each year, one or more working groups meet in Lyon, France, to consider the scientific information available on exposures with carcinogenic potential. Since 1977 these working groups have designated exposures according to their potential for carcinogenicity in humans.³ The designations are:

Group 1 — Carcinogenic to humans;

Group 2A — Probably carcinogenic to humans;

Group 2B — Possibly carcinogenic to humans;

Group 3 — Not classifiable as to its carcinogenicity to humans;

Group 4 — Probably not carcinogenic to humans.

Each of the above designa tions takes into account information from epidemiologic studies in humans and, to a lesser extent, from experimental studies in laboratory animals. In order for an agent to be judged as belonging in Group 1, as a rule there must be sufficient evidence of information relating to humans based on epidemiologic studies. Group 2A indicates limited evidence for humans and sufficient evidence for animals. Group 2B indicates inadequate evidence in humans together with sufficient evidence in animals or limited evidence in both humans and animals. Group 4 implies lack of carcinogenicity in both human and animals. Finally, all other exposures fall into Group 3.

In 1979, the first comprehensive list of exposures judged to cause human cancer was published by IARC.³ Table 1 is based on this report and lists the substances present or processes that occur in the workplace, together with the cancer caused and the year that each was suggested by the IARC to be a carcinogen. Several of these agents or exposures had been suspected to be carcinogenic before 1900, and only two were initially identified after 1970. It should be noted that the year an agent was first thought to be a carcinogen is somewhat arbitrary; other dates have been suggested.⁴

Table 1. Agents judged by IARC to be human carcinogens (a)

(a) Substances updated in later reviews are not included.
(b) Years in parentheses refer to work environment rather than to specific compound.
(c) Hematopoietic and lymphatic cancers.

——————————————————————

Many of the exposures listed in Table 1 are associated with an increased cancer risk of 10 times or greater. The estimated risk increase depends, of course, on the level and duration of exposure to a working population as well as on the availability of reliable data. For example, shoe workers exposed to benzene-containing glues may have had a leukemia risk that was 10 to 100 times the risk in the general population; however, no epidemiologic data are available. In contrast, the Pliofilm cohort in Ohio, on which the primary epidemiologic study of people occupationally exposed to benzene was based, had an increased risk of leukemia of about 3.4.⁵

Many of the ag ents listed in Table 1 are relatively weak human carcinogens with an increased risk no higher than about five times. For some of these agents, for instance, ethylene oxide (EtO), epidemiologic evidence by itself was not sufficient to categorize them as Group 1 carcinogens; the mechanism of action of EtO was also used to make the final judgment.

One agent—ionizing radiation—has not been evaluated by IARC but is clearly a human carcinogen.⁶ Watch-dial painters in the early part of this century used radium containing paint to make the dial glow in the dark. The painters, mostly women, would twirl the brush on their tongue to make a point on the brush. The radium that they ingested via this route was concentrated in their bones, and osteogenic sarcoma was an unfortunate but relatively common result. Also, radiologists who practiced in the early 20th century had an elevated rate of leukemia as a result of their occupational exposure to radiation.

Table 2 lists occupational exposures that to date have been classified by IARC as Group 2A—probable human carcinogens. These agents represent a mixture of chemical compounds and manufacturing processes or more general exposures. Each has been brought to our attention only recently as being possibly carcinogenic. All of these agents, if they are in fact carcinogenic to humans, are relatively weak causes of the cancers listed. There is legitimate question as to whether sufficient epidemiologic information can be collected to provide an unambiguous judgment of causality.

Table 2. Agents judged by IARC from 1979 through 1995 to be probable human carcinogens

(a) Hematopoietic and lymphatic cancers

Improvements in industrial hygiene
Almost all of the agents listed in Table 1 are substances or processes to which humans have been exposed in the workplace for many decades and, in some cases, even centuries. People have been exposed to these carcinogens at relatively high levels even in the second half of this century. Before 1900, little attention was given to the possibility that an agent in the workplace caused cancer or indeed any chronic disease. As long as a worker could function in an environment that may have been filled with dusts, mists, and gases, the primary concern was economic gain, often for the employer. During the first half of the 20th century, attention gradually was drawn to the possibility that exposure in the workplace may lead to chronic disease and cancer;⁷ however, a lack of acute toxicity still served as the indicator that the workplace was safe. Only since World War II have industrial hygiene and epidemiology developed into professions that consider the possible long-term dangers of exposures in the workplace. In the United States, the Occupational Safety and Health Act in 1970 brought official notice to workers and management that continuing vigilance was needed to prevent cancer among industrial workers.⁸

It is hoped that no worker will develop cancer because of current occupational exposures in developed countries. Ideally, known carcinogens or carcinogenic processes have been contained so that worker exposure is kept to a minimum, and no new carcinogens will be introduced into the workplace. The combination of reduction of exposure to known carcinogens and minimization of introduction of new carcinogens will likely lead to the inability of epidemiologists to identify new carcinogens from classic case-control or cohort studies, because the risks involved will be very low. Rather, continued vigilance or surveillance will be based on monitoring of physiologic change in workers or by evaluation of markers of genetic damage in cells.

Summary Points

Suggestions

Suggested Further Reading

References